Axis deviation estimation device

Abstract

The present invention relates to a radar device. An object information acquisition unit (31) acquires object information including the object distance between the radar device and a reflecting object and the object azimuth angle at which the reflecting object exists. A roadside object extraction unit (33) extracts roadside object information related to the roadside object from the object information. When the orientation of the radar device in a reference state is used as a mounting reference direction and the actual orientation of the radar device is used as an actual mounting direction, an axis deviation angle estimation unit (35) estimates a vertical axis deviation angle of the actual mounting direction in a direction perpendicular to the mounting reference direction based on the roadside object information.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This international application claims the benefit of priority based on Japanese Patent Application No. 2020-47819 filed with the Japan Patent Office on March 18, 2020, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to a technique for estimating an axial offset of a radar device. Background Art

[0004] Conventional vehicle-mounted radar devices can experience a shift in the central axis of the radar beam due to changes in their setup, such as due to various factors. This shift in the central axis can reduce the radar device's detection accuracy.

[0005] As a countermeasure, for example, Patent Document 1 below discloses a technique for estimating the angle of vertical axis offset of a radar device (ie, vertical axis offset) by utilizing the phenomenon that the reception intensity of reflected waves from a road surface near a vehicle reaches a maximum.

[0006] Patent Document 1: Japanese Patent No. 6321448

[0007] As a result of detailed studies by the inventors regarding the above-mentioned technology, they discovered the following problems.

[0008] In the above-mentioned technology, since the angle of vertical axis deviation (i.e., vertical axis deviation angle) is estimated by using the reception intensity of the reflected wave on the road surface, it is not easy to estimate the vertical axis deviation angle with good accuracy when the radar beam is facing upward (i.e., when the sensor is facing upward).

[0009] That is, when the radar beam is deflected upward, there are cases where the reflected waves from the road surface cannot be sufficiently received. In such a case, it is difficult to detect the deflection based on the reflected waves. Summary of the Invention

[0010] One aspect of the present disclosure is to provide a technology capable of estimating the vertical axis offset angle of a radar device with high accuracy.

[0011] An axial deviation estimation device according to one aspect of the present disclosure relates to an axial deviation estimation device for estimating an axial deviation of a radar device mounted on a mobile object.

[0012] The axial deviation estimation device includes an object information acquisition unit, a roadside object extraction unit, and an axial deviation angle estimation unit.

[0013] The object information acquisition unit is configured to repeatedly acquire object information including an object distance, which is the distance between the radar device and a reflecting object corresponding to a reflection point of a radar wave detected by the radar device, and an object azimuth, which is the azimuth at which the reflecting object exists.

[0014] The roadside object extraction unit is configured to extract roadside object information related to the roadside object from the object information. Specifically, the unit is configured to extract the roadside object information from the object information based on predetermined extraction conditions. The roadside object information represents information about reflection points on the roadside object located to the side of the road on which the mobile vehicle is traveling, at a position higher than the road, and along the direction in which the road extends, in accordance with predetermined conditions.

[0015] The axis deviation angle estimation unit is configured to estimate the vertical axis deviation angle representing the deviation angle in the vertical direction of the actual mounting direction relative to the mounting reference direction based on roadside object information including information of multiple reflection points, with the orientation of the radar device when the radar device is mounted in a reference state as the mounting reference direction and the actual orientation of the radar device as the mounting actual direction.

[0016] With this configuration, in one aspect of the present disclosure, it is possible to easily extract roadside object information, such as the position of a roadside object located along the travel path, from object information related to the reflecting object obtained by operating the radar device. Because the roadside object is located along the travel path at a higher position than the travel path and in accordance with prescribed conditions, even if, for example, the radar beam is displaced upward, the reflected wave from the roadside object is easier to detect than the reflected wave from the road surface.

[0017] That is, even if the radar device is tilted upward, the reflected waves from roadside objects are easier to detect than the reflected waves from the road surface. In addition, roadside objects are easier to detect even at a distance.

[0018] Therefore, in one aspect of the present disclosure, by utilizing roadside objects having such characteristics, the vertical axis offset of the radar device can be estimated with high accuracy based on roadside object information obtained by reflected waves from the roadside objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a block diagram showing a vehicle control system including the axial misalignment estimation device according to the first embodiment.

[0020] Figure 2 This is an explanatory diagram illustrating the horizontal irradiation range of radar waves.

[0021] Figure 3 This is an explanatory diagram illustrating the vertical irradiation range of radar waves.

[0022] Figure 4This is a block diagram functionally showing the axial misalignment estimation device according to the first embodiment.

[0023] Figure 5 This is an explanatory diagram for explaining the vertical axis offset angle and the roll angle.

[0024] Figure 6 This is an explanatory diagram for explaining the axis deviation of the radar device.

[0025] Figure 7 This is an explanatory diagram illustrating the arrangement of road guardrails and the like on a two-dimensional plane.

[0026] Figure 8 This is an explanatory diagram illustrating the guardrail and the arrangement of its reflection points in the vertical direction.

[0027] Figure 9 This is an explanatory diagram showing the relationship between the vertical axis offset angle, the arrangement of reflection points, and the approximate straight line.

[0028] Figure 10 This is a flowchart showing the main routine of the axis offset estimation process.

[0029] Figure 11 This is a flowchart showing the roadside property candidate point extraction process.

[0030] Figure 12 This is a flowchart showing the roadside object point cloud extraction process.

[0031] Figure 13 This is a flowchart showing the direct axis offset angle estimation process.

[0032] Figure 14 This is an explanatory diagram illustrating inflection points in the reflection point cloud.

[0033] Figure 15 This is an explanatory diagram illustrating the relationship between vehicle coordinates, device coordinates, and the vertical axis offset angle.

[0034] Figure 16 This is a flowchart showing the processing in the second embodiment.

[0035] Figure 17 This is a flowchart showing the processing in the third embodiment.

[0036] Figure 18 This is a flowchart showing the processing in the fourth embodiment. DETAILED DESCRIPTION

[0037] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0038] [1. First embodiment]

[0039] [1-1. Overall structure]

[0040] First, the overall configuration of a vehicle control system including the axial misalignment estimation device according to the first embodiment will be described.

[0041] Figure 1 The vehicle control system 1 shown is a system installed in a vehicle VH, which is a mobile object. The vehicle control system 1 primarily includes a radar device 3 and a control device 5. It may also include an onboard angle adjustment device 7, an onboard sensor group 9, an axle misalignment notification device 11, and an assist actuator 13. Hereinafter, the vehicle VH on which the vehicle control system 1 is installed will also be referred to as the host vehicle VH. Furthermore, the vehicle width direction of the host vehicle VH will be referred to as the horizontal direction, and the vehicle height direction will be referred to as the vertical direction.

[0042] like Figure 2 as well as Figure 3 As shown, radar device 3 is mounted on the front side of the host vehicle VH and radiates radar waves in the direction forward of the host vehicle VH (i.e., in the direction of travel). Specifically, radar device 3 radiates radar waves within a predetermined angular range Ra in the horizontal direction and a predetermined angular range Rb in the vertical direction in front of the host vehicle VH. Radar device 3 receives reflected waves from the radiated radar waves and generates reflection point information (i.e., object information) regarding the point (i.e., reflecting object) at which the radar waves were reflected.

[0043] Furthermore, radar device 3 may be a so-called millimeter-wave radar that uses electromagnetic waves in the millimeter-wave band as radar waves, a lidar that uses lasers as radar waves, or a sonar that uses sound waves as radar waves. In short, the antenna unit that transmits and receives radar waves is configured to detect the direction of arrival of reflected waves in both the horizontal and vertical directions. The antenna unit may also include an array antenna arranged in the horizontal and vertical directions.

[0044] The radar device 3 is installed so that the beam of radar waves (i.e., radar beam) emitted by it is aligned with the beam direction in the front-to-rear direction of the host vehicle VH, and therefore aligned with the direction of travel. Furthermore, the radar device 3 is used to detect various objects (i.e., target objects) in front of the host vehicle VH. The beam direction is the direction along the central axis CA of the radar beam. When the radar device 3 is installed in the correct position (i.e., the reference position), the beam direction generally coincides with the direction of travel.

[0045] The reflection point information generated by the radar device 3 includes at least the azimuth of the reflection point and the distance to the reflection point (i.e., the distance between the radar device 3 and the reflection point). Furthermore, the radar device 3 may be configured to detect the relative speed of the reflection point relative to the vehicle VH and the reception intensity (i.e., received power) of the reflected radar wave from the reflection point. The reflection point information may also include the relative speed and reception intensity of the reflection point.

[0046] like Figure 2 as well as Figure 3 As shown, the azimuth of the reflection point is an angle calculated relative to the direction along the central axis CA of the radar beam, i.e., the beam direction. Specifically, it is at least one of the horizontal angle (hereinafter referred to as the horizontal angle) Hor and the vertical angle (hereinafter referred to as the vertical angle) Ver at the reflection point. Both the vertical angle Ver and the horizontal angle Hor are included in the reflection point information as information indicating the azimuth of the reflection point.

[0047] The radar device 3 uses, for example, the FMCW method to alternately transmit radar waves in an uplink modulation interval and a downlink modulation interval at a preset modulation cycle, and receives reflected radar waves. FMCW is an abbreviation for Frequency Modulated Continuous Wave.

[0048] As described above, the radar device 3 detects the horizontal angle Hor and vertical angle Ver as the azimuth of the reflection point, the distance from the reflection point, the relative speed to the reflection point, and the reception intensity of the received radar wave as reflection point information at each modulation cycle.

[0049] The mounting angle adjustment device 7 includes a motor and a gear mounted on the radar device 3. The mounting angle adjustment device 7 rotates the motor in response to a drive signal output from the control device 5. The motor's rotational force is thereby transmitted to the gear, allowing the radar device 3 to rotate about an axis extending in the horizontal direction and an axis extending in the vertical direction.

[0050] Therefore, for example, by making the radar device 3 move in the direction of arrow A along the vertical plane (for example, referring to Figure 5 ) can be rotated to adjust the offset angle in the vertical direction of the radar device 3.

[0051] The vehicle-mounted sensor group 9 is at least one sensor mounted on the vehicle VH to detect the state of the vehicle VH. The vehicle-mounted sensor group 9 may also include a vehicle speed sensor. The vehicle speed sensor is a sensor that detects the vehicle speed based on the rotation of the wheels. Figure 1 As shown, the vehicle-mounted sensor group 9 may include a camera 15 such as a CCD camera. The camera 15 captures the same range as the irradiation range of the radar wave of the radar device 3.

[0052] The onboard sensor group 9 may also include an acceleration sensor. The acceleration sensor detects the acceleration of the vehicle VH. Furthermore, the onboard sensor group 9 may also include a yaw rate sensor. The yaw rate sensor detects the rate of change of the yaw angle, which represents the slope of the vehicle VH's traveling direction relative to the direction in front of the vehicle VH. Furthermore, the onboard sensor group 9 may also include a steering angle sensor. The steering angle sensor detects the steering angle of the steering wheel.

[0053] Furthermore, the vehicle-mounted sensor group 9 may include a navigation device 17 having map information. The navigation device 17 may also be a device that detects the position of the vehicle VH based on GPS signals or the like and associates the position of the vehicle VH with the map information. The map information may also include various road-related information such as roadside objects, such as guardrails (hereinafter referred to as guardrails) 41 for vehicles (see, for example, Figure 7 )’s location information.

[0054] The axial misalignment notification device 11 is a sound output device installed in the vehicle cabin, and outputs a warning sound to the passengers of the host vehicle VH. Alternatively, an audio device or the like included in the assist execution unit 13 may be used as the axial misalignment notification device 11.

[0055] The assistance execution unit 13 controls various in-vehicle devices and performs prescribed driving assistance based on the results of object detection processing (described later) performed by the control device 5. The various in-vehicle devices that are controlled may include monitors that display images and audio equipment that outputs warning sounds and guidance voices. Furthermore, control devices for controlling the internal combustion engine, power transmission mechanism, braking mechanism, and the like of the vehicle VH may also be included.

[0056] The control device 5 includes a microcomputer 29 that includes a CPU 19 and a semiconductor memory (hereinafter referred to as "memory") 27, such as a ROM 21, a RAM 23, and a flash memory 25. The various functions of the control device 5 are realized by the CPU 19 executing a program stored on a non-migratable physical recording medium. In this example, the memory 27 corresponds to the non-migratable physical recording medium that stores the program. Furthermore, the execution of the program executes the method corresponding to the program. The control device 5 may include a single microcomputer 29 or multiple microcomputers 29.

[0057] like Figure 4 As shown, the control device 5 includes the functions of an object information acquisition unit 31 , a roadside object extraction unit 33 , and an axis deviation angle estimation unit 35 , and functions as an axis deviation estimation device.

[0058] The object information acquisition unit 31 repeatedly acquires reflection point information (ie, object information) including the azimuth angle of the reflection point (ie, object azimuth angle) and the distance of the reflection point (ie, object distance).

[0059] The roadside object extraction unit 33 extracts roadside object information from the aforementioned reflection point information based on predetermined extraction conditions described later. This roadside object information represents information about reflection points on roadside objects (e.g., guardrails 41) located to the side of the road (i.e., lane) on which the vehicle VH is traveling, at a higher position than the road surface and along the direction in which the road extends, according to predetermined conditions (e.g., the same height). Furthermore, the roadside object information includes, for example, information about the positions of reflection points where radar waves are reflected by roadside objects.

[0060] The axis deviation angle estimation unit 35 estimates the vertical axis deviation angle based on the roadside object information. Specifically, the vertical axis deviation angle, which represents the deviation angle in the vertical direction of the actual mounting direction relative to the estimated mounting reference direction, is estimated based on the roadside object information including information on multiple reflection points, assuming that the orientation of the radar device 3 when mounted in a reference state (i.e., reference position) is defined as the mounting reference direction and the actual orientation of the radar device 3 is defined as the actual mounting direction.

[0061] Here, the so-called mounting reference direction is the direction of the radar device 3 when the radar device 3 is mounted at the original installation position (i.e., the pre-set position), that is, the reference position. In the first embodiment, the mounting reference direction is, for example, Figure 2 as well as Figure 3 The direction of the X-axis (i.e., Xc) shown is consistent, and when the radar device 3 is mounted in the reference position, there is no axis deviation in the radar device 3. In addition, the front direction of the radar device 3 is the orientation of the radar device 3 (i.e., the reference orientation), and the front direction of the vehicle VH is the mounting reference direction.

[0062] [1-2. Radar device axis deviation]

[0063] Next, the axis offset of the radar device 3 will be described.

[0064] The axis offset of the radar device 3 refers to the offset of the coordinate axis of the radar device 3 when the radar device 3 is actually installed on the vehicle VH relative to the coordinate axis of the radar device 3 when the radar device 3 is correctly installed on the vehicle VH.

[0065] The axis offset of the radar device 3 includes an axis offset around the device coordinate axis and an axis offset in the height direction. Here, among the axis offsets around the device coordinate axis, the vertical axis offset will be mainly described.

[0066] (a) Coordinate axis

[0067] First, the coordinate axes of the radar device 3 and the coordinate axes of the host vehicle VH will be described.

[0068] like Figure 5As shown, the coordinate axes of the radar device 3 are the vertical axis Zs extending vertically along the radar device 3, the horizontal axis Ys extending horizontally along the radar device 3, and the front-rear axis Xs extending front-to-back along the radar device 3 when the radar device 3 is mounted on the host vehicle VH. The vertical axis Zs, the horizontal axis Ys, and the front-rear axis Xs are mutually orthogonal. In this first embodiment, in which the radar device 3 is installed in front of the host vehicle VH, the front-rear axis Xs coincides with the central axis CA of the radar beam. In other words, the orientation of the radar device 3 coincides with the front-rear axis Xs.

[0069] Furthermore, the coordinates in the radar device 3 (ie, device-system coordinates) are constituted by the up-down axis Zs, the left-right axis Ys, and the front-back axis Xs.

[0070] Meanwhile, the coordinate axes of the host vehicle VH are the vertical axis Zc, which extends in the vertical direction; the horizontal axis Yc, which extends in the horizontal direction; and the travel direction axis Xc, which extends in the direction of travel of the host vehicle VH. The vertical axis Zc, the horizontal axis Yc, and the travel direction axis Xc are orthogonal to each other.

[0071] Furthermore, the vertical axis Zc, the horizontal axis Yc, and the traveling direction axis Xc constitute coordinates in the host vehicle VH (ie, vehicle coordinates).

[0072] Furthermore, in the first embodiment, as described above, when the radar device 3 is properly mounted on the vehicle VH, the central axis CA coincides with the direction of travel of the vehicle VH. In other words, the coordinate axes of the radar device 3 and the coordinate axes of the vehicle VH coincide in their respective directions. For example, in an initial state, such as when shipped from the factory, the radar device 3 is properly mounted on the vehicle VH, that is, mounted in a predetermined position.

[0073] (b) Axis offset around the device coordinate axis

[0074] Next, the axis offset around the device coordinate axis will be described.

[0075] After the initial state, the vehicle VH may experience axial deviation around the device coordinate axis. This axial deviation includes vertical axis deviation and roll axis deviation. The axial deviation angle represents the magnitude of this axial deviation as an angle.

[0076] Among them, such as Figure 5As shown in the left figure, vertical axis offset refers to a state in which an offset occurs between the vertical axis Zs, the coordinate axis of the radar device 3, and the vertical axis Zc, the coordinate axis of the host vehicle VH. The offset angle during this vertical axis offset is referred to as the vertical axis offset angle θp. The vertical axis offset angle θp is the so-called pitch angle θp, representing the offset angle of the radar device 3's coordinate axis relative to the horizontal axis Yc of the host vehicle VH. Specifically, the vertical axis offset angle θp is the offset angle when an offset occurs around the horizontal axis Yc of the host vehicle VH, and thus around the left-right axis Ys of the radar device 3.

[0077] In addition, as from Figure 5 As can be seen from the left figure, the vertical axis offset angle θp can also be an angle that represents the magnitude of the offset between the front-rear axis Xs as the coordinate axis of the radar device 3 and the traveling direction axis Xc as the coordinate axis of the vehicle VH.

[0078] Here, based on Figure 6 , the vertical axis offset angle is described in more detail.

[0079] Figure 6 This figure shows a state where the radar beam of radar device 3 is deflected (i.e., deflected in the vertical direction) in the ZX plane, a plane perpendicular to the travel axis Xc. When no deflection occurs, the central axis CA of the radar beam is identical to the travel axis Xc.

[0080] like Figure 6 As shown, when the mounting reference direction of the radar device 3 is consistent with the traveling direction of the vehicle VH, when the actual orientation of the radar device 3, that is, the actual mounting direction, is set as the beam direction, in the vertical direction, the angle between the traveling direction and the beam direction is the vertical axis offset angle θp.

[0081] That is, when the central axis CA of the radar beam of the radar device 3 deviates from the reference traveling direction to the actual beam direction in the figure due to the radar device 3 rotating in the direction of arrow A, the deviation angle is the vertical axis deviation angle θp.

[0082] In addition, as mentioned above Figure 5 As shown in the right figure of FIG, roll axis offset refers to a state in which an offset occurs between the left-right axis Ys, which is the coordinate axis of the radar device 3, and the horizontal axis Yc, which is the coordinate axis of the host vehicle VH. The axis offset angle during such roll axis offset is referred to as the roll angle θr.

[0083] [1-3. Principle]

[0084] Next, the principle of estimating the vertical axis deviation angle using roadside objects as in the first embodiment will be described.

[0085] (a) For example, Figure 7as well as Figure 8 As shown, as a roadside object, a case where there is a guardrail 41 on the side of the road in the width direction and arranged in the direction in which the road extends so as to protrude upward from the road surface is described as an example. Figure 7 The left and right directions are the width directions of the road. Figure 7 The up-down direction is the direction in which the road extends, that is, the direction in which the vehicle VH travels.

[0086] like Figure 8 As shown, such guardrails 41 are generally arranged at the same height along the direction in which the road extends. Specifically, a plurality of posts 43 are arranged in a row along the direction in which the road extends, and rod-shaped or plate-shaped cross members 45 are fixed to laterally connect the posts 43 (e.g., adjacent posts 43).

[0087] That is, the posts 43 and cross members 45 are generally arranged at a constant height, so that the upper end of the guardrail 41 extends almost horizontally along the road. In addition, the entire guardrail 41 also extends almost horizontally on the road surface in a strip shape on a vertical plane (i.e., with a predetermined vertical width).

[0088] Therefore, when the radar device 3 of the vehicle VH irradiates a radar beam forward, the radar beam is reflected by the road surface and the guardrail 41, and the reflected wave is received by the radar device 3. Therefore, based on the reflected wave, the road surface and the guardrail 41 are detected as reflection points (i.e., reflecting objects).

[0089] When a radar beam is actually irradiated from the radar device 3 toward the guardrail 41 and its reflected waves are detected, the intensity of the reflected waves from the upper ends of the pillars 43 and the upper ends of the cross members 45 is relatively high, making it easy to detect the reflection points at the upper ends of the pillars 43 and the upper ends of the cross members 45. Furthermore, reflection points at locations other than the upper ends of the pillars 43 and the upper ends of the cross members 45 in the guardrail 41 can also be detected.

[0090] Therefore, when a guardrail 41 is arranged along the road, multiple reflection points corresponding to the guardrail 41 are detected in a strip-shaped range along the traveling direction of the vehicle VH. In particular, reflection points corresponding to the upper ends of the pillars 43 and the upper ends of the cross members 45 are detected in a narrow, substantially linear range.

[0091] Therefore, as described in detail later, the slope of the reflection point cloud when there is a vertical axis offset can be obtained based on the arrangement of a plurality of reflection points (ie, reflection point cloud) detected in a strip-shaped range corresponding to the guardrail 41 .

[0092] In addition, Figure 8 In FIG. 1 , for ease of understanding, a straight line is shown that connects the reflection points at the upper ends of the pillars 43 and the upper ends of the horizontal members 45 to represent the arrangement of the reflection point clouds.

[0093] (b) Next, based on Figure 9 , the relationship between the vertical axis offset angle θ and the reflection point cloud is explained.

[0094] like Figure 9 As shown in (B), when no vertical axis offset occurs in the radar device 3 (i.e., when the center axis CA is horizontal), the configuration of multiple reflection points detected by the radar device 3 on the vertical plane is also close to horizontal, as shown in the graph on the right side of the figure.

[0095] also, Figure 9 The graph to the right of shows the positions of each reflection point in the three-dimensional device coordinate system when projected onto the Z-X plane along the left-right axis Ys (i.e., the projected reflection point). Furthermore, the straight lines in each graph are approximate lines KL obtained by least-squares approximating the multiple projected reflection points. Below, the projected reflection points may be simply referred to as reflection points.

[0096] Therefore, if Figure 9 As shown in the right graph of (B), when the approximate straight line KL is determined to be horizontal based on the detection result of the radar device 3, it can be determined that no vertical axis offset has occurred.

[0097] However, if Figure 9 As shown in (A), when the center axis CA of the radar beam of the radar device 3 (i.e., the direction of the radar device 3) is offset downward, the center axis CA of the radar beam moves further away from the upper end of the column 43 toward the direction of travel shown by Xc (i.e., farther).

[0098] In addition, in the above Figure 8 , when the central axis CA of the radar beam is offset downward relative to the traveling direction axis Xc, the distance between the central axis CA and the upper end of the guardrail 41 increases as one moves to the right side of the figure.

[0099] Therefore, if Figure 9 As shown in the graph on the right side of (A), the more the reflection points are arranged, the higher they are, and therefore the slope β of the approximate straight line KL has a positive value. In addition, the greater the absolute value of the slope β of the approximate straight line KL, the greater the absolute value of the downward vertical axis deviation angle θp of the radar device 3. That is, as shown in FIG. Figure 9 As can be seen from (A) and the like, the absolute value of the angle corresponding to the slope β of the approximate straight line KL (ie, the tilt angle βk) is the same as the absolute value of the vertical axis deviation angle θp of the radar device 3, but their signs are opposite.

[0100] Therefore, if Figure 9As shown in the graph on the right side of (A), when the slope β of the approximate straight line KL is determined based on the detection results of the radar device 3 (i.e., a positive value of β), it can be determined that a downward vertical axis offset occurs at a vertical axis offset angle θp corresponding to the slope β. In this case, the tilt angle βk is a positive value, and the vertical axis offset angle θp is a negative value.

[0101] On the contrary, if Figure 9 As shown in (C), when the radar device 3 is oriented upward, the arrangement of the reflection points decreases as it moves toward the direction of travel (i.e., toward the right side of the figure), as shown in the graph on the right side of the figure. In this case, the slope β of the approximate straight line KL has a negative value in the device coordinate system.

[0102] Therefore, if Figure 9 As shown in the graph on the right side of (C), when the slope β of the approximate straight line KL is determined based on the detection results of the radar device 3 (i.e., a negative value of β), it can be determined that an upward vertical axis offset occurs at a vertical axis offset angle θp corresponding to the slope β. In this case, the tilt angle βk is a negative value, and the vertical axis offset angle θp is a positive value.

[0103] Thus, the vertical axis offset of the radar device 3 , that is, the vertical axis offset angle θp, can be determined from the slope of the arrangement of reflection points on the ZX plane, that is, the slope β of the approximate straight line KL.

[0104] [1-4. Processing]

[0105] Next, the processing performed by the control device will be described.

[0106] (a) Main routine for axis offset estimation processing

[0107] First, use Figure 10 The entire shaft offset estimation process (ie, the main routine) executed by the control device 5 will be described with reference to the flowchart of FIG.

[0108] This axis offset estimation process is a process for estimating the vertical axis offset angle θp, and is started when the ignition switch is turned on.

[0109] When this process is started, in step (hereinafter referred to as S) 100, the control device 5 performs a process of detecting an object in front of the vehicle VH using the radar device 3. This object detection process is a so-called target object detection process, and is a well-known process as described in, for example, Japanese Patent No. 6321448 mentioned above, so a detailed description thereof will be omitted.

[0110] Here, the object (ie, the object marker) corresponds to the reflection point indicated by the reflection point information. At this stage, the reflection point includes not only the road surface but also roadside objects such as the guardrail 41 .

[0111] Specifically, in S100, reflection point information is acquired from the radar device 3. Reflection point information refers to information about each of the multiple reflection points detected by the radar device 3 mounted on the vehicle VH. The reflection point information includes at least the horizontal and vertical angles representing the azimuth of the reflection point, and the distance between the radar device 3 and the reflection point. Furthermore, the control device 5 acquires various detection results, including the vehicle speed Cm, from the onboard sensor group 9.

[0112] In the next S110 , a roadside object candidate extraction process is executed. As will be described in detail later, this roadside object candidate extraction process is a process for extracting reflection points that are candidates for roadside objects (ie, roadside object candidate points) from a plurality of reflection points obtained by the radar device 3 .

[0113] In the next S120 , a roadside object point cloud extraction process is executed. As will be described in detail later, this roadside object point cloud extraction process is a process for further extracting point clouds (i.e., roadside object point clouds) with a high probability of being roadside objects from the plurality of roadside object candidate points obtained in S110 .

[0114] In the next S130 , a vertical axis offset angle estimation process is executed. As will be described in detail later, this vertical axis offset angle estimation process is a process for estimating the vertical axis offset angle θp of the radar device 3 based on the roadside object point cloud obtained in the above S120 .

[0115] In the next S140, it is determined whether the vertical axis offset angle θp estimated in the above S130 requires adjustment by the mounting angle adjustment device 7. If the determination is affirmative, the process proceeds to S150, while if the determination is negative, the process proceeds to S180.

[0116] Specifically, when the vertical axis deviation angle θp of the radar device 3 is greater than a predetermined threshold angle, it is determined that adjustment is necessary and the process proceeds to S150 . On the other hand, when it is less than the threshold angle, the process proceeds to S180 .

[0117] In S150, it is determined whether the vertical axis offset angle θp is within the adjustable range of the mounting angle adjustment device 7. If the determination is affirmative, the process proceeds to S170, while if the determination is negative, the process proceeds to S160.

[0118] In S170 , since the vertical axis offset angle θp is within the adjustable range, the axis offset adjustment process is executed. That is, the mounting angle adjustment device 7 is controlled to adjust the vertical axis offset angle θp to zero.

[0119] Specifically, the radar device 3 is rotated around the left-right axis Ys by an amount corresponding to the vertical axis offset angle θp, with the left-right axis Ys of the radar device 3 as the center, so that the orientation of the radar device 3 becomes the mounting reference direction, and the process proceeds to S180.

[0120] On the other hand, in S160, since the vertical axis offset angle θp is outside the adjustable range, the vertical axis offset angle θp is not adjusted. Instead, diagnostic information indicating that an axis offset has occurred in the radar device 3 (i.e., an axis offset diagnosis) is output to the axis offset notification device 11, and the process proceeds to S180. Furthermore, the axis offset notification device 11 may also output a warning sound based on the axis offset diagnosis.

[0121] In S180, for example, it is determined whether the present process should be terminated based on whether the ignition switch is off. If an affirmative determination is made, the present process is temporarily terminated, while if a negative determination is made, the process returns to the above-mentioned S100.

[0122] (b) Roadside object candidate point extraction and processing

[0123] Next, use Figure 11 The roadside object candidate point extraction process executed by the control device 5 will be described with reference to the flowchart of FIG.

[0124] This process is the above Figure 10 The process of S110 is for extracting a candidate reflection point of a roadside object (i.e., a roadside object candidate point) from the plurality of reflection points obtained by the radar device 3. The candidate reflection point extracted here is a reliable point as the reflection point of the guardrail 41 described above.

[0125] In the following description, a roadside object is taken as an example of the guardrail 41 , but there is also a case where the guardrail 41 is simply referred to as a roadside object.

[0126] First, in Figure 11 In S200, it is determined whether the "determination condition based on distance" is established (ie, whether it is satisfied). If an affirmative determination is made here, the process proceeds to S210, while if a negative determination is made, the process proceeds to S260.

[0127] For example, regarding the reflection point to be determined in the traveling direction of the vehicle VH, it is determined whether or not the condition that the reflection point exists within a range exceeding 2 m and less than 100 m from the vehicle VH is satisfied.

[0128] In S210 , it is determined whether the “determination condition based on the lateral position” is satisfied. If the determination is affirmative, the process proceeds to S220 , while if the determination is negative, the process proceeds to S260 .

[0129] For example, determine whether the condition that "when the vehicle VH is traveling on a left-hand traffic road (e.g., a two-lane road), the reflection point exists on the left side of the vehicle VH in the direction of travel within a range of more than 2m and less than 8m from the vehicle" is met.

[0130] Furthermore, for example, when the vehicle VH is traveling on a single-lane road, it may be determined whether a reflection point exists on the right side of the vehicle VH in a range exceeding 2 m and less than 8 m from the vehicle.

[0131] That is, in this S210 , it is determined whether the reflection point is in a range in the lateral direction of the vehicle VH where the guardrail 41 as a roadside structure is likely to exist.

[0132] In S220 , it is determined whether the “determination condition based on relative speed” is satisfied. If the determination is affirmative, the process proceeds to S230 , while if the determination is negative, the process proceeds to S260 .

[0133] Specifically, since guardrail 41 is a stationary object, the condition that the speed of the reflection point relative to the host vehicle VH (i.e., the relative speed) corresponds to the speed of the host vehicle VH representing the stationary object (i.e., the host vehicle speed Cm) is determined to be satisfied. Furthermore, if the host vehicle speed Cm is positive, the detected relative speed is negative.

[0134] Furthermore, the relative speed can be determined based on whether the absolute value of the relative speed is within a predetermined error range of ±Δ centered around the absolute value of the vehicle speed Cm.

[0135] In S230, it is determined whether the "determination condition based on the running state of the host vehicle VH (ie, the host vehicle state)" is satisfied. If the determination is affirmative, the process proceeds to S240, while if the determination is negative, the process proceeds to S260.

[0136] For example, when the vehicle VH is traveling straight and the acceleration is constant, the detection accuracy of the reflection point can be considered high. Therefore, here, based on the information from the vehicle sensor group 9, it is determined whether the vehicle state is a stable state of steady travel.

[0137] For example, when the vehicle VH is traveling, if the yaw angle detected by the yaw rate sensor or the steering wheel rotation angle detected by the steering angle sensor is below a predetermined value, it may be determined that the vehicle is traveling in a straight line. Alternatively, if the acceleration detected by the acceleration sensor is below a predetermined value, it may be determined that the acceleration is constant.

[0138] Furthermore, when determining that the vehicle is traveling in a straight line or that the acceleration is constant, it may be determined that the vehicle is traveling in a straight line and that the acceleration is constant if the error is within a predetermined error range.

[0139] In S240 , it is determined whether the “determination condition by the camera 15 ” is satisfied. If the determination is affirmative, the process proceeds to S250 , while if the determination is negative, the process proceeds to S260 .

[0140] For example, the image captured by the camera 15 may be processed using a known image processing method, and based on this image, it may be determined whether the image of the object located at the reflection point is highly likely to be the guardrail 41. Furthermore, methods for detecting the guardrail 41 based on the image captured by the camera 15 are known, for example, as described in Japanese Patent Application Laid-Open No. 2011-118753.

[0141] In S250, regarding the reflection point of the determination object, since a positive judgment is made in all the above steps S200 to S240, the reflection point is stored in the memory 27 as a roadside object candidate point with a high possibility of being a reflection point of the guardrail 41, and this processing is temporarily ended.

[0142] On the other hand, in S250 , since a negative determination is made in any of S200 to S240 , the reflection point is stored in the memory 27 as a non-roadside object with a low probability of being the guardrail 41 , and the present process is temporarily terminated.

[0143] Furthermore, since the above-mentioned processes of S200 to S260 are performed on all the reflection points obtained by the above-mentioned object detection process, all the reflection points are classified as either roadside object candidates or non-roadside objects.

[0144] (c) Roadside object point cloud extraction and processing

[0145] Next, use Figure 12 The flowchart of the roadside object point cloud extraction process performed by the control device 5 is explained.

[0146] This process is the above Figure 10 The processing of S120 is used to Figure 11 The roadside object candidate point extraction process is used to extract the multiple roadside object candidate points obtained by the roadside object candidate point extraction process for calculating the vertical axis offset angle θp. In addition, the roadside object point cloud is composed of multiple reflection points.

[0147] First, in Figure 12 In S300, candidate point clustering processing is performed. That is, a plurality of roadside object candidate points are clustered (ie, classified).

[0148] For example, multiple roadside object candidate points, or reflection points, are divided into multiple (e.g., six) clusters using the well-known k-means method. Furthermore, each reflection point is three-dimensional data with X, Y, and Z coordinates in vehicle coordinates, and clustering is performed using the X, Y coordinates of each reflection point.

[0149] In the next S310 , it is determined whether the “longitudinal distance determination condition of the roadside object point cloud (ie, point cloud)” is satisfied. If the determination is positive, the process proceeds to S320 , while if the determination is negative, the process proceeds to S350 .

[0150] That is, in each of the divided clusters, it is determined whether the longitudinal distance determination condition is satisfied for all roadside object candidate points (ie, roadside object point cloud) included in each cluster.

[0151] Specifically, for example, for each roadside object point cloud corresponding to each cluster, that is, for all reflection points in each roadside object point cloud, a determination is made as to whether the length of the vehicle VH in the travel direction, i.e., the depth direction, lies above a certain range. Specifically, for all reflection points in each cluster to be determined, a determination is made as to whether the value obtained by subtracting the distance to the vehicle VH (the minimum value) from the maximum depth distance of the reflection points from the vehicle VH is greater than a predetermined threshold.

[0152] Through the determination in S310 , clusters that satisfy the vertical distance determination condition of the point cloud can be extracted from all clusters. In other words, clusters having reflection points that satisfy the vertical distance determination condition of the point cloud can be extracted from all clusters.

[0153] Here, for each cluster, the longitudinal distance determination condition for that cluster is considered satisfied when all reflection points satisfy the aforementioned distance condition. However, the longitudinal distance determination condition for that cluster may also be considered satisfied when a predetermined proportion or more of the reflection points satisfy the aforementioned distance condition. This also applies to the following determination conditions.

[0154] In S320 , it is determined whether the “cross-distance determination condition of the point cloud” is satisfied. If the determination is positive, the process proceeds to S330 , while if the determination is negative, the process proceeds to S350 .

[0155] That is, for a cluster in which an affirmative determination is made in the above-mentioned S310 , it is determined whether the lateral distance determination condition is satisfied for all the reflection points of the roadside object point cloud of the cluster.

[0156] Specifically, for example, for all reflection points in the cluster, it is determined whether the length of the vehicle VH in the left-right direction, i.e., the width direction, is within a predetermined range or less. Specifically, for all reflection points, it is determined whether the value obtained by subtracting the closest distance to the vehicle VH (the minimum value) from the farthest distance to the vehicle VH (the maximum value) in the width direction is greater than a predetermined threshold.

[0157] Through the determination in S320 , clusters satisfying the lateral distance determination condition of the point cloud can be further extracted from clusters satisfying the longitudinal distance determination condition of the point cloud.

[0158] In S330, it is determined whether the "determination condition of the lateral position" is satisfied. If the determination is affirmative, the process proceeds to S340, while if the determination is negative, the process proceeds to S350.

[0159] That is, for the cluster determined to be affirmative in the above-mentioned S320 , it is determined whether the determination condition for the horizontal position is satisfied.

[0160] Specifically, it is determined whether the point cloud of the cluster to be determined is the point cloud on the innermost side in the left-right direction of the vehicle VH.

[0161] For example, when the right side of the vehicle is considered positive in left-hand traffic, it is determined whether the lateral position of the point cloud is positive (ie, the right side of the vehicle) and the position closest to the vehicle.

[0162] In addition, when the right side of the vehicle is considered positive in left-hand traffic, it is determined whether the lateral position of the point cloud is negative (ie, the left side of the vehicle) and the position closest to the vehicle.

[0163] In S340 , since all the above S310 to S330 are determined to be positive, the point cloud of the selected cluster is regarded as a point cloud representing reflection points of roadside objects (ie, roadside object point cloud) and stored in the memory 27 , temporarily terminating the present process.

[0164] On the other hand, in S350, since a negative judgment is made in any one of the above S310 to S330, the point cloud of the cluster in which the negative judgment is made is regarded as a point cloud that does not represent the reflection points of roadside objects (i.e., non-roadside object point cloud), and this processing is temporarily terminated.

[0165] The determination processing of S310 to S330 is performed to extract reliable reflection points of roadside objects such as the fingerprint guardrail 41 .

[0166] (d) Vertical axis offset angle estimation processing

[0167] Next, use Figure 13 The vertical axis deviation angle estimation process executed by the control device 5 is described with reference to the flowchart of FIG.

[0168] This process is the above Figure 10 The process of S130 is used to Figure 12 The vertical axis offset angle θp is calculated based on the roadside object point cloud (i.e., reflection point cloud) obtained by the roadside object point cloud extraction process.

[0169] First, in S400, for each roadside object point (i.e., the reflection point corresponding to the roadside object point) in the roadside object point cloud obtained by the above-mentioned roadside object point cloud extraction processing, the coordinates of the position of each roadside object point (i.e., the device system coordinates) are calculated based on the distance and azimuth contained in the reflection point information corresponding to each roadside object point.

[0170] The so-called device coordinates are three-dimensional coordinates based on the coordinate axes of the radar device 3, that is, coordinates expressed by (Xs, Ys, Zs). Figure 10 The object detection process is obtained.

[0171] That is, the control device 5 calculates the coordinates (Xs, Ys, Zs) of all roadside object points (ie, reflection points) in the roadside object point cloud as the device system coordinates, and stores them in the memory 27 .

[0172] In the next step S410, it is determined whether the deviation determination condition of the position of each roadside object point in the roadside object point cloud (i.e., the roadside object position) is satisfied. If the determination is positive, the process is temporarily terminated. On the other hand, if the determination is negative, the process proceeds to S420.

[0173] The deviation determination condition determines whether the roadside object point cloud (i.e., multiple reflection points) is dispersed to such an extent that it is difficult to approximate the above-mentioned approximate straight line KL in the Z-X plane of the device coordinate system (i.e., whether the degree of deviation is greater than a predetermined value). For example, the correlation coefficient of the multiple reflection points in the Z-X plane can be used as this determination condition.

[0174] That is, in the first embodiment, since the vertical axis offset angle θp is estimated using the approximate straight line KL, cases with large deviations are excluded and states with small deviations that allow the approximate straight line KL to be estimated are extracted.

[0175] In S420, since the deviation was determined to be small in S410, the least squares method is used to calculate the equation (1) for the approximate straight line KL for all reflection points in the roadside object point cloud. Specifically, the following approximate straight line KL is calculated on the Z-X plane of the device coordinate system. The slope of equation (1) is β, and C is the intercept.

[0176] Zs=βXs+C··(1)

[0177] In the next S430, it is determined whether the inflection point determination condition is satisfied. If the determination is positive, the process proceeds to S440. On the other hand, if the determination is negative, the present process is temporarily terminated.

[0178] like Figure 14As shown, the inflection point determination condition is a condition for determining whether the arrangement of the plurality of roadside object points (ie, the plurality of reflection points) on the ZX plane is substantially straight as a whole in the device-system coordinates.

[0179] For example, Figure 14 As shown, the approximate straight line KL is calculated for all reflection points in the roadside object point cloud, and straight lines SL are drawn between adjacent reflection points. The angle at which the approximate straight line KL intersects each straight line SL can also be calculated. If the angle of intersection is greater than a specified value, it is determined that the inflection point determination condition is not met (i.e., an inflection point exists). Furthermore, the two reflection points used to draw the straight line SL do not need to be adjacent reflection points, but rather the two reflection points that are at least a specified distance away from each other can be used.

[0180] That is, in this first embodiment, when a roadside object such as the guardrail 41 is continuous along the road in a constant state, for example, at a constant height, since the vertical axis offset angle θp is estimated, it is determined here whether the roadside object is continuous in such a state.

[0181] also, Figure 14 The above figure shows an example of a roadside point cloud without inflection points. Figure 14 The figure below shows an example of a roadside object point cloud with an inflection point. "Inflection points" means that the arrangement of multiple reflection points is not a straight line, but rather bends midway.

[0182] In S440, the angle corresponding to the slope β of equation (1) representing the approximate straight line KL (i.e., the inclination angle βk) is obtained, and the vertical axis offset angle θp is obtained by reversing the positive and negative values ​​of the angle, and this process is temporarily terminated.

[0183] In this way, the vertical axis deviation angle θp of the radar device 3 can be obtained.

[0184] In addition, Figure 15 The relationship between the device-based coordinates and the vehicle-based coordinates is shown in FIG. Here, since the vertical axis offset angle θp is positive when the radar device 3 is offset toward the upward axis, for example, the front-back axis Xs of the device-based coordinates rotates counterclockwise relative to the travel-direction axis Xc of the vehicle-based coordinates by an amount corresponding to the vertical axis offset angle θp.

[0185] Therefore, in the vehicle coordinate system, the straight line indicating the direction of the radar device 3 , that is, the central axis CA, can be expressed by the following equation (2): C is the intercept.

[0186] Zc=θpXc+C··(2)

[0187] [1-5. Effect]

[0188] In the above-mentioned first embodiment, the following effects can be obtained.

[0189] (1a) The first embodiment includes an object information acquisition unit 31 , a roadside object extraction unit 33 , and an axis deviation angle estimation unit 35 .

[0190] With this configuration, in the first embodiment, it is possible to easily extract roadside object information, such as the position of the reflection point of a roadside object such as a guardrail 41 located along the roadway, from the reflection object information corresponding to the reflection point of the radar wave obtained by driving the radar device 3. For example, since the guardrail 41 is located at a constant height along the road at a higher position than the road surface, even if the radar beam is displaced upward, the reflected wave from the guardrail 41 is easier to detect than the reflected wave from the road surface.

[0191] That is, even if the direction of the radar device 3 is shifted upward, the reflected wave from the guardrail 41 is easier to detect than the reflected wave from the road surface. In addition, the guardrail 41 is easier to detect even at a distance.

[0192] Therefore, in the first embodiment, by utilizing roadside objects such as the guardrail 41 having such characteristics, the vertical axis deviation angle θp of the radar device 3 can be estimated with high accuracy based on roadside object information obtained by reflected waves from the roadside object.

[0193] (1b) In the first embodiment, based on the aforementioned roadside object information, the arrangement of multiple reflection points of roadside objects such as the guardrail 41 on a vertical plane along the travel direction of the host vehicle VH is approximated by a straight line. The vertical axis offset angle θp can then be estimated using this approximate straight line KL.

[0194] For example, guardrail 41 is arranged in a strip shape at a constant height along a vertical plane. Specifically, guardrail 41 is arranged in a strip shape so as to extend continuously along the road at a constant height and parallel to the road surface. Consequently, the distribution of multiple radar wave reflection points on the vertical plane forms a nearly strip-shaped distribution with a slope corresponding to the vertical axis offset angle θp. Therefore, the vertical axis offset angle θp can be estimated with high accuracy based on the approximate straight line KL derived from this strip-shaped distribution of reflection points.

[0195] (1c) In the first embodiment, when the distribution of the reflection points on the vertical plane obtained based on the reflection points of the guardrail 41 is in a state with an inflection point, that is, when the configuration of multiple reflection points is approximated by a straight line, the vertical axis offset angle θp is not estimated.

[0196] That is, since the vertical axis offset angle θp is estimated when the conditions for accurately estimating the vertical axis offset angle θp are satisfied, the vertical axis offset angle θp can be obtained with high accuracy.

[0197] (1d) In the first embodiment, when the deviation of the positions of the plurality of reflection points of the roadside object on the vertical plane is greater than or equal to a predetermined value based on the roadside object information, the vertical axis offset angle θp is not estimated.

[0198] That is, since the vertical axis offset angle θp is estimated when the conditions for accurately estimating the vertical axis offset angle θp are satisfied, the vertical axis offset angle θp can be obtained with high accuracy.

[0199] (1e) In the first embodiment, when the host vehicle VH is traveling straight, the vertical axis deviation angle θp is estimated.

[0200] That is, since the vertical axis offset angle θp is estimated when the conditions for accurately estimating the vertical axis offset angle θp are satisfied, the vertical axis offset angle θp can be stably obtained with high accuracy.

[0201] [1-6. Correspondence between statements]

[0202] In the relationship between this first embodiment and the present disclosure, the vehicle VH corresponds to the mobile body, the radar device 3 corresponds to the radar device, the control device 5 corresponds to the axis offset estimation device, the object information acquisition unit 31 corresponds to the object information acquisition unit, the roadside object extraction unit 33 corresponds to the roadside object extraction unit, and the axis offset angle estimation unit 35 corresponds to the axis offset angle estimation unit.

[0203] [2. Second embodiment]

[0204] Since the basic structure of the second embodiment is the same as that of the first embodiment, the following description will focus on the differences from the first embodiment.

[0205] In the second embodiment, the axis deviation angle estimation unit 35 is configured to estimate the vertical axis deviation angle θp by weighting the reflection information representing roadside object information that is located farther than a predetermined distance from the vehicle VH.

[0206] For example, in the case where reflection points corresponding to multiple roadside objects are detected, Figure 16 As shown, in S500, the control device 5 determines whether the reflection point is a reflection point located in a distant range that is at least a predetermined distance from the vehicle VH. If the reflection point is a distant reflection point, the control device 5 increases the number of reflection points, for example, by doubling the number, in S510.

[0207] Therefore, when the approximate straight line KL is obtained by the least square method for a plurality of reflection points, the approximate straight line KL can be obtained based on the reflection points after the addition of distant reflection points (ie, after reset).

[0208] also, Figure 16 The processing can be done for example Figure 13 Therefore, the position of the reflection point of the roadside object can be reset.

[0209] Specifically, near the radar device 3, various noises are easily superimposed on the reflected waves, making it more difficult to obtain accurate information such as the location of the reflection point compared to locations farther from the radar device 3. Therefore, in this second embodiment, information about reflection points farther from the radar device 3 is prioritized and weighted.

[0210] Therefore, since the more accurate arrangement of the reflection points is known, a more accurate approximate straight line KL with less error can be obtained. Therefore, a more accurate vertical axis offset angle θp can be obtained based on the more accurate approximate straight line KL.

[0211] Furthermore, in the second embodiment as well, the same effects as those of the first embodiment can be obtained.

[0212] [3. Third embodiment]

[0213] Since the basic structure of the third embodiment is the same as that of the first embodiment, the following description will focus on the differences from the first embodiment.

[0214] The third embodiment is configured to use the map information when extracting reflective object information representing the roadside object information, when map information representing the roadside object information includes information on the position of roadside objects such as guardrails 41 .

[0215] For example, Figure 17 As shown, in the control device 5 , in S600 , it is determined whether the map used by the navigation device 17 is a map in which the positions of roadside objects such as the guardrail 41 are recorded.

[0216] Then, if the map contains the locations of roadside objects, then in S610, based on the map information and the position information of the host vehicle VH, it is determined whether a roadside object such as a guardrail 41 is installed along the road on which the host vehicle VH is traveling. Then, if the road is a road with roadside objects, in S620, information on the position of the roadside object relative to the host vehicle VH is obtained, such as the range of the roadside object in a two-dimensional plane.

[0217] Furthermore, in the case of a road without roadside objects, since there are no roadside objects required for estimating the axis offset, various processes required for estimating the axis offset do not need to be performed.

[0218] above Figure 17 The processing can be done for example Figure 11 This is performed before the roadside object candidate point extraction process shown. Furthermore, for example, information on the placement range of roadside objects obtained from map information can be utilized before or after any of the processes S200 to S240. Specifically, a process for narrowing down the number of roadside object candidate points is provided before or after the processes S200 to S240, and information on the placement range of roadside objects obtained from the map information can be utilized as a judgment condition for this process.

[0219] In this way, the above-described processing allows the position and range of roadside objects to be determined based on map information. Therefore, when the radar device 3 actually detects roadside objects, this map information can be used to accurately extract the roadside objects. Consequently, the vertical axis offset angle θp can be more accurately estimated.

[0220] Furthermore, in the third embodiment as well, the same effects as those of the first embodiment can be obtained.

[0221] [4. Fourth embodiment]

[0222] Since the basic structure of the fourth embodiment is the same as that of the first embodiment, the following description will focus on the differences from the first embodiment.

[0223] like Figure 1 As shown, the fourth embodiment is configured as a radar device 3, which includes a front radar device 3a for detecting the traveling direction of the vehicle VH, that is, objects in front (i.e., reflecting objects), and a side radar device 3b for detecting objects on the side of the vehicle VH (i.e., reflecting objects).

[0224] The fourth embodiment is configured to estimate the vertical axis deviation angle θp when a roadside object can be detected by the front radar device 3 a and the side radar device 3 b .

[0225] For example, Figure 18 As shown, in the control device 5, when it is determined in S700 that a roadside object can be detected by the front radar device 3a and in S710 that a roadside object can be detected by the side radar device 3b, in S720, the estimation of the vertical axis offset angle θp is allowed.

[0226] In addition, for example, after the roadside object candidate extraction process or the roadside object point cloud extraction process is performed by each of the radar devices 3a and 3b, Figure 18 processing.

[0227] Furthermore, ultimately, the vertical axis deviation angle θp can be estimated using the reflection point information obtained by the front radar device 3 a .

[0228] This makes it possible to reliably determine roadside objects, and thereby obtain the vertical axis deviation angle θp with high accuracy.

[0229] Furthermore, in the fourth embodiment, the same effects as those of the first embodiment can be obtained.

[0230] [5. Other embodiments]

[0231] As mentioned above, although embodiment of this disclosure was described, this disclosure is not limited to the said embodiment, Various deformation|transformation and implementation are possible.

[0232] (5a) In the present disclosure, a radar device is not limited to one capable of detecting roadside objects in front of (i.e., in front of) the vehicle. In the present disclosure, a radar device capable of detecting roadside objects in any of the following directions: rearward, frontward (e.g., diagonally in front of the left or diagonally in front of the right), or lateral (e.g., left or right) to the vehicle. In other words, any radar device capable of detecting roadside objects such as guardrails is not particularly limited.

[0233] Furthermore, at least two or more of the above-mentioned radar devices may be combined. For example, the vertical axis deviation angle may be estimated using reflective object information from a radar device capable of detecting roadside objects.

[0234] (5b) In the present disclosure, various radar systems utilizing various methods, such as the 2FCW method, the FCM method, and the pulse method, can be employed as radar devices in addition to the aforementioned FMCW method. 2FCW stands for 2-Frequency Modulated Continuous Wave, and FCM stands for Fast-Chirp Modulation.

[0235] (5c) In each of the above-described embodiments, data obtained by the radar device is transmitted to a control device (e.g., an axis deviation estimation device) and processed (e.g., axis deviation estimation processing). However, the radar device itself may process the data (e.g., axis deviation estimation processing performed by the axis deviation estimation device). Furthermore, data may be processed by each sensor in the on-vehicle sensor group, or data obtained by each sensor may be transmitted to a control device, and various processing may be performed by the control device.

[0236] (5d) Roadside structures other than guardrails may include multiple curbstones arranged along the road, multiple lane-dividing columns, and the like. Furthermore, guardrails may include various types of vehicle guardrails, such as guardrails, tubular guardrails, cable guardrails, and box-beam guardrails, as well as barriers for pedestrians and bicycles.

[0237] Furthermore, as roadside structures, for example, a plurality of structures, such as the aforementioned plurality of curbstones and a plurality of pillars, or a single structure can be used. For example, various guardrails arranged continuously and integrally across a long distance in the direction in which the road extends, or sidewalls made of concrete or the like can be used.

[0238] (5e) The control device and method thereof disclosed herein may also be implemented using a dedicated computer provided by a processor and a memory programmed to execute one or more functions embodied by a computer program.

[0239] Alternatively, the control device and method described in the present disclosure may be implemented using a dedicated computer provided by a processor composed of one or more dedicated hardware logic circuits.

[0240] Alternatively, the control device and method described in the present disclosure may be implemented using one or more dedicated computers formed by a combination of a processor and a memory programmed to execute one or more functions, and a processor composed of one or more hardware logic circuits.

[0241] In addition, the computer program may be stored as a computer-executable instruction in a computer-readable non-transferable physical recording medium. The method for implementing the functions of each component included in the control device does not necessarily need to include software, and all functions may be implemented using one or more hardware components.

[0242] (5f) It is also possible to implement multiple functions of one component in the above-mentioned embodiment by multiple components, or to implement one function of one component by multiple components. In addition, it is also possible to implement multiple functions of multiple components by one component, or to implement one function implemented by multiple components by one component. In addition, it is also possible to omit a part of the structure of the above-mentioned embodiment. In addition, it is also possible to add or replace at least a part of the structure of the above-mentioned embodiment with the structure of other above-mentioned embodiments.

[0243] (5g) In addition to the above-mentioned control device, the present disclosure can also be implemented in various ways, such as a system with the control device as a component, a program that enables a computer to function as the control device, a non-migrating physical recording medium such as a semiconductor memory that records the program, a control method, etc.

Claims

1. An axial offset estimation device for estimating an axial offset of a radar device mounted on a mobile object, comprising: an object information acquiring unit configured to repeatedly acquire object information including an object distance (i.e., a distance between the radar device and a reflecting object corresponding to a reflection point of a radar wave detected by the radar device) and an object azimuth (i.e., an azimuth angle) at which the reflecting object exists; a roadside object extraction unit configured to extract roadside object information from the object information based on predetermined extraction conditions, the roadside object information representing information of the reflection point on a roadside object located at a higher position than the travel path and in a direction extending along the travel path in accordance with predetermined conditions, to the side of the travel path on which the mobile body is traveling; as well as The axis deviation angle estimation unit is configured to estimate a vertical axis deviation angle indicating a deviation angle in the vertical direction of the actual mounting direction relative to the above-mentioned reference mounting direction based on the above-mentioned roadside object information including information of the plurality of above-mentioned reflection points, when the orientation of the above-mentioned radar device when the above-mentioned radar device is mounted in a reference state is used as the mounting reference direction and the actual orientation of the above-mentioned radar device is used as the actual mounting direction. The axis deviation angle estimation unit is configured to estimate the vertical axis deviation angle by weighting the roadside object information with an emphasis on information on the roadside object located farther than a predetermined distance in the roadside object information.

2. The axial deviation estimation device according to claim 1, wherein: The axis deviation angle estimation unit is configured to estimate the vertical axis deviation angle using the roadside object information of the roadside object having a constant position in the height direction.

3. The axial deviation estimation device according to claim 1 or 2, wherein: The axis deviation angle estimation unit is configured to approximate the configuration of the multiple reflection points of the roadside object along the traveling direction of the moving body on a vertical plane with a straight line based on the roadside object information, and use the straight line to estimate the vertical axis deviation angle.

4. The axial deviation estimation device according to claim 1 or 2, wherein: The axis offset angle estimation unit is configured as follows: Based on the roadside object information, the arrangement of the plurality of reflection points of the roadside object along the traveling direction of the moving body on a vertical plane is approximated by a first straight line, and, When the configuration of the multiple reflection points of the roadside objects on the vertical plane is divided into multiple areas along the driving direction, and each of the divisions is approximated by a second straight line, the vertical axis offset angle is not estimated when the difference in slope between the first straight line and the second straight line is greater than a specified value.

5. The axis deviation estimation device according to claim 1 or 2, wherein: The axis deviation angle estimation unit is configured not to estimate the vertical axis deviation angle when a deviation in the positions of the plurality of reflection points of the roadside object on the vertical plane is greater than a predetermined value based on the roadside object information.

6. The axial deviation estimation device according to claim 1 or 2, wherein: The vertical axis deviation angle is estimated when the moving object is traveling in a straight line.

7. The axial deviation estimation device according to claim 1 or 2, wherein: When the map information indicating the travel route and its surroundings includes information on the position of the roadside object, the map information is used when extracting the roadside object information.

8. The axial deviation estimation device according to claim 1 or 2, wherein: The configuration is such that, when the radar device is provided with a front radar device for detecting the above-mentioned reflecting object in the front, i.e., the traveling direction of the above-mentioned moving body, and a side radar device for detecting the above-mentioned reflecting object on the side of the above-mentioned moving body, the estimation of the above-mentioned vertical axis deviation angle is implemented when the above-mentioned roadside object can be detected by the above-mentioned front radar device and the above-mentioned side radar device.

9. A method for estimating an axis offset of a radar device mounted on a mobile object, comprising: repeatedly acquiring object information including an object distance (i.e., a distance between the radar device and a reflecting object corresponding to a reflection point of a radar wave detected by the radar device) and an object azimuth (i.e., an azimuth angle) at which the reflecting object exists; Extracting roadside object information from the object information based on predetermined extraction conditions, the roadside object information representing information of the reflection point on a roadside object arranged in accordance with predetermined conditions at a position higher than the travel path and in a direction extending along the travel path to the side of the travel path of the mobile body; as well as When the orientation of the radar device when the radar device is mounted in a reference state is used as the mounting reference direction and the actual orientation of the radar device is used as the actual mounting direction, a vertical axis offset angle indicating an offset angle in the vertical direction of the actual mounting direction relative to the mounting reference direction is estimated based on the roadside object information including information of the plurality of reflection points. The vertical axis deviation angle is estimated by giving weight to the roadside object information by giving importance to the roadside object information located at a position farther than a predetermined distance, thereby estimating the vertical axis deviation angle.

10. A computer program product comprising a computer program which, when executed by a processor, implements the method according to claim 9.

Images